lids. Dimly foreshadowed in this method of development are the more complete modern types of insect metamorphosis, which have their morphological origin, as we now know, in a doubling of most of the rudiments of the organs in the embryo. On hatching, one set of these rudiments develops immediately into the larval body, while the other set remains in abeyance in the form of minute germinal centers, or histoblasts, from which the body of the adult will be fashioned during the quiescent pupal stage. The higher insects are therefore beautiful examples of double personality, much more perfect examples of this phenomenon, in fact, than any that has been discovered in man. The larval insect is, if I may be permitted to lapse for a moment into anthropomorphism, a sluggish, greedy, self-centered creature, while the adult is industrious, abstemious and highly altruistic, concentrating its activities on reproduction and the dissemination of the species. Unlike ourselves, who are Mr. Hydes and Dr. Jekylls in varying degrees, for brief alternating periods in our lives, or even simultaneously, the youthful insect sows its wild oats with a vengeance as a glutton or even as an assassin and then experiences a change of heart and reforms for good and all.
Parasitism must have been very easily grafted on to such a sharply dichotomic method of development as that of the holometabolous insects, for the larvæ of the predators are already much inclined to sloth and gluttony when the food supply is abundant, and comparatively little modification would be required to convert them into parasites. But the same peculiarities of metamorphosis have also made the holometabolic insects ideal hosts. We have already seen that insects, as a rule, are themselves not only parasitic during larval life, but also prefer larvæ as hosts. It is not improbable that this is the primitive, and that parasitism on the egg, pupa or adult is a secondary, or derivative condition. The real secret of both host and parasite being larvae lies in the peculiar significance of anabolism in this stage. The host accumulates great quantities of fats and proteids as a so-called "fat-body," which is of little or no immediate use to the organism itself, but is stored up to be utilized during metamorphosis. This fat body may, therefore, be devoured by the parasite and converted into its own fat-body without seriously injuring the host. Furthermore, the fact that the parasite, too, stores up its food in the form of a fat-body instead of at once turning it over to its gonads and becoming reproductive, accounts for the striking differences between the insect parasite, on the one hand, and the tape-worm and Sacculina, on the other. The few exceptions among insects, such as the female Strepsiptera, in which the food taken by the larval parasite from its host is soon turned over to the gonads and used for reproduction, leads to a permanent parasitism resembling that of the tapeworms or the adult Sacculina. The larva is at once arrested in its development and begins to reproduce by pasdogenesis. We may con-
